Move `library` calls to the top of the Rmd document

[elray1]

It's helpful to be up-front about what the package requirements are

[LucieContamin]

If I understand correctly, it seems that the manuscript needs some specific version of the packages we use, we might need to add that information.

[eahowerton]

Are we targeting Journal of Statistical Software? I was perusing some of the papers to see what the conventions are, and it seems many do not include the library calls explicitly in the main text of the paper. @elray1 would you suggesting hiding these at the top? Or working them into the text somehow, just earlier on?

We can always do something slightly different for the vignette too.

[elray1]

Yeah, I think the Journal of Statistical Software is a good target.

I have a couple of slightly conflicting thoughts:

• For the manuscript, I agree with your idea that it makes sense to me to follow the conventions you see there in terms of how much detail is/isn't shown, so suppressing the library calls could make sense in general...
• except, perhaps for any hubverse packages we're using? It seems like it might be good to be clear that hubverse provides a few different packages, and although the main focus of this paper is hubEnsembles, we are using some functions from other hubverse packages along the way. That might suggest a few clear library calls to those hubverse packages at the top, and/or some text describing this somewhere?

To some extent, as long as we do something reasonable I think this won't be a determinant of whether the work is accepted, and we can take some feedback about this from the reviewers/editors too.